The objective of this research is to obtain a clearer estimate of the size and extent of the surviving population of cochlear nerve fibers in the ears of deaf patients who might be considered as canidates for an auditory prosthesis designed to bypass the organ of Corti and deliver electrical stimulation directly to the nerve fibers. We propose to study temporal bones obtained at autopsy from patients known to have been deafened by noise exposure, ototoxic drugs, aging, or viral disease, and to measure the extent of loss of both sensory and neural structures, including the organ of Corti, the myelinated dendrites in the osseous spiral lamina, the cells of the spiral ganglion and the axons in the modiolus. We have also been promised the temporal bones which might possibly be available from the patients presently fitted with implants. The techniques of microdissection and phase-contrast examination of surface preparations will be employed, followed by sectioning of the decalcified modiolus embedded in celloidin. Monkeys and chinchillas trained for behavioral audiometry will be used as animal models. After measurement of their normal thresholds for pure tones they will be deafened by ototoxic drugs, noise exposure, or combinations of the two that are known to produce extensive injury to cochlear structures. When their residual hearing, if any, has been measured audiometrically, the animals will be kept for varying periods up to 2 years, then sacrificed for post-mortem study of the inner ears to determine the extent of cochlear injury and degeneration, and the degree to which the cochlear nerve fibers have survived. Information from this research should be helpful in predicting the likelihood of long-term success for the use of implanted cochlear prostheses in various forms of acquired deafness. Similarly Dalmatian dogs will be used as animal models for studying cochlear nerve degeneration in hereditary and congenital deafness.